INDEX

Explanations

events or situations that are planned or organized

oai_token-act-pair · gpt-3.5-turbo

instances of the word "staged" and its variations, indicating planned or organized events

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iability

-0.68

iosis

-0.66

tymology

-0.66

imb

-0.66

icient

-0.66

etheless

-0.66

uyomi

-0.65

scribe

-0.65

conom

-0.65

inoa

-0.64

POSITIVE LOGITS

 staging

0.88

 staged

0.87

ificial

0.70

 rall

0.69

 roller

0.67

ivals

0.67

URN

0.66

 barric

0.66

 Assass

0.66

querade

0.65

Activations Density 0.016%

events or situations that are planned or organized

instances of the word "staged" and its variations, indicating planned or organized events

No Comments

No Known Activations